AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
LLM Stats
MMStar
MMStar is an elite vision-indispensable multimodal benchmark comprising 1,500 challenge samples meticulously selected by humans to evaluate 6 core capabilities and 18 detailed axes. The benchmark addresses issues of visual content unnecessity and unintentional data leakage in existing multimodal evaluations.
Evaluation context and evidence
- Source
- LLM Stats
- Publisher
- Shanghai AI Laboratory
- Benchmark release
- 2024-04-09
- Collected
- 2026-08-17
- Modality
- multimodal
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 24
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| Qwen3.6 Plus | 0.833 | Not recorded | Not recorded | 2026-04-02 · model_announcement | LLM Stats |
| Qwen3.5-122B-A10B | 0.829 | Not recorded | Not recorded | 2026-02-24 · model_announcement | LLM Stats |
| Qwen3.5-35B-A3B | 0.819 | Not recorded | Not recorded | 2026-02-24 · model_announcement | LLM Stats |
| Qwen3.6-27B | 0.814 | Not recorded | Not recorded | 2026-04-21 · model_announcement | LLM Stats |
| Qwen3.5-27B | 0.81 | Not recorded | Not recorded | 2026-02-24 · model_announcement | LLM Stats |
| Qwen3 VL 32B Thinking | 0.794 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 235B A22B Thinking | 0.787 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 235B A22B Instruct | 0.784 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 32B Instruct | 0.777 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 30B A3B Thinking | 0.755 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 8B Thinking | 0.753 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 4B Thinking | 0.732 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 30B A3B Instruct | 0.721 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen3 VL 8B Instruct | 0.709 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen2.5 VL 72B Instruct | 0.708 | Not recorded | Not recorded | 2025-01-26 · model_announcement | LLM Stats |
| Qwen3 VL 4B Instruct | 0.698 | Not recorded | Not recorded | 2025-09-22 · model_announcement | LLM Stats |
| Qwen2.5 VL 32B Instruct | 0.695 | Not recorded | Not recorded | 2025-02-28 · model_announcement | LLM Stats |
| Qwen2.5-Omni-7B | 0.64 | Not recorded | Not recorded | 2025-03-27 · model_announcement | LLM Stats |
| Qwen2.5 VL 7B Instruct | 0.639 | Not recorded | Not recorded | 2025-01-26 · model_announcement | LLM Stats |
| LFM2.5-VL-3B | 0.633 | Not recorded | Not recorded | 2026-08-12 · model_announcement | LLM Stats |
| DeepSeek VL2 | 0.613 | Not recorded | Not recorded | 2024-12-13 · model_announcement | LLM Stats |
| DeepSeek VL2 Small | 0.57 | Not recorded | Not recorded | 2024-12-13 · model_announcement | LLM Stats |
| North Micro Vision Instruct | 0.518 | Not recorded | Not recorded | 2026-08-12 · model_announcement | LLM Stats |
| DeepSeek VL2 Tiny | 0.459 | Not recorded | Not recorded | 2024-12-13 · model_announcement | LLM Stats |